linux-kernel - Re: [PATCH 1/4] x86/entry: Make paranoid

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 21 Dec 2021 10:22:48 +0800
From:   Lai Jiangshan <laijs@...ux.alibaba.com>
To:     Borislav Petkov <bp@...en8.de>,
        Lai Jiangshan <jiangshanlai@...il.com>
Cc:     linux-kernel@...r.kernel.org, x86@...nel.org,
        Andy Lutomirski <luto@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        "H. Peter Anvin" <hpa@...or.com>
Subject: Re: [PATCH 1/4] x86/entry: Make paranoid_exit() callable



On 2021/12/21 02:56, Borislav Petkov wrote:
> On Mon, Dec 13, 2021 at 11:03:37PM +0800, Lai Jiangshan wrote:
>> From: Lai Jiangshan <laijs@...ux.alibaba.com>
>>
>> Move the last JMP out of paranoid_exit() and make it callable.
>>
>> It will allow asm_exc_nmi() to call it and avoid duplicated code.
>>
>> No functional change intended.
>>
>> Signed-off-by: Lai Jiangshan <laijs@...ux.alibaba.com>
>> ---
>>   arch/x86/entry/entry_64.S | 18 +++++++++++-------
>>   1 file changed, 11 insertions(+), 7 deletions(-)
>>
>> diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
>> index 44dadea935f7..3dc3cec03425 100644
>> --- a/arch/x86/entry/entry_64.S
>> +++ b/arch/x86/entry/entry_64.S
>> @@ -439,7 +439,8 @@ SYM_CODE_START(\asmsym)
>>   
>>   	call	\cfunc
>>   
>> -	jmp	paranoid_exit
>> +	call	paranoid_exit
>> +	jmp	restore_regs_and_return_to_kernel
> 
> I guess but I don't like the glueing of the CALL to paranoid_exit with
> the JMP to the restore_regs_and_return_to_kernel label. That reads
> to me as, "if you're calling paranoid_exit() you must jump to the
> restore_regs_and_return_to_kernel label but not always."
> 
> So I'm thinking you should leave the jump to that label inside
> paranoid_exit() and use its %rbx argument to control when to jump to it
> and when not.

Hello

Thank you for reviewing it.

This patch is a part of the job converting some ASM code to C code.  The
changelog of it in the original big patchset reads:

   Allow paranoid_exit() to be re-written in C later and also allow
   asm_exc_nmi() to call it to avoid duplicated code.

And this smaller patchset doesn't include the work of converting ASM code,
so I removed "Allow paranoid_exit() to be re-written in C later" because I
thought "allowing asm_exc_nmi() to call it and avoiding duplicated code" is
enough to justify the value of this change.

When paranoid_exit() is ready to be converted to C, it can't have jump to
any label that is not in paranoid_exit() itself.

I'm sorry to not have put all of the intents in the changelog:

-It will allow asm_exc_nmi() to call it and avoid duplicated code.
+It will allow asm_exc_nmi() to call it and avoid duplicated code and allow
+future work to convert paranoid_exit() to C code.

Thanks
Lai